Our long term goal is to elucidate the molecular mechanism of HMGB1 A box (A box), a DNA binding domain of the molecule, as an antagonist of HMGB1. Severe sepsis is the 3rd leading cause of death nationwide, and does not currently have sufficient medical treatment, finding relevant therapeutic targets and understanding the mechanisms underlying this pathology are highly significant. HMGB1 is a necessary and sufficient mediator of severe sepsis and has been proven to be involved in both sepsis progression, organ damage and in sterile injury. In structure-function analyses, we observed that A box (amino acids 1-85 on HMGB1 protein), has an antagonistic effect to HMGB1. A box specifically inhibits HMGB1-induced cytokine release in macrophage cultures. Administration of A box rescues mice from lethal endotoxemia or sepsis caused by cecal ligation and puncture, a clinically relevant sepsis model. Following our first publication in 2004, several reports confirmed that A box acts as an antagonist of HMGB1 and extended our observations by showing beneficial effects of A box in several models of infection and sterile injury where excessive amount of HMGB1 release was observed. But the mechanisms underlying A box-mediated antagonistic effects of HMGB1 remain unknown. In aim 1 of this proposal, we will synthesize segments of A box (over-lapping 20-mer peptides) and make point mutation mutants (cysteines at positions of 23 and 45 replaced by serine) of A box and assess the inhibition of these segments/mutants on HMGB1-mediated cytokine responses in macrophages; to start to identify the minimum sequence and/or critical region of A box required as an antagonist of HMGB1 in aim 2, we will investigate mechanism of A box action by examining the interaction of A box, and derivatives identified from above, with HMGB1 and with HMGB1 receptor TLR4/MD2; in aim 3, we will examine the therapeutic efficacy of these effective A box derivatives in clinical relevant animal models of endotoxemia and sepsis induced by cecal perforation. We hypothesize that A box segment(s) binds to HMGB1 and inhibits HMGB1 further binding to MD2 in the TLR4/MD2 complex, consequently inhibits HMGB1-induced toxicity in sepsis. We further hypothesize that cysteines on positions 23 and 45 of A box are critical for the antagonistic activity of A box to HMGB1. Results of this research will contribute both to the scientific effort of elucidating this important mechanism for A box action, as well as to yield new directions in developing therapeutics for sepsis, by using HMGB1 antagonists as drug targets.